Jobe J B, White A A, Kelley C L, Mingay D J, Sanchez M J, Loftus E F
Office of Research and Methodology, National Center for Health Statistics, Hyattsville, MD 20782.
Milbank Q. 1990;68(2):171-89.
Complex questions in health surveys place heavy cognitive demands on respondents, prompting researchers to appraise how specific cognitive interventions may improve the accuracy of people's answers. Investigators in one experiment asked participants to recall visits to medical providers in forward, backward, or no particular order, and matched results with providers' records. "Free" recall proved marginally superior to forward or backward ordering, although overall respondents underreported the number of visits by 20 percent; participants' gender and self-reported health status, among other factors, also affected quality of recall. The experiment lends support to contentions that the methods of cognitive science applied to survey research better the accuracy of population survey data.
健康调查中的复杂问题对受访者提出了很高的认知要求，促使研究人员评估特定的认知干预措施如何提高人们回答的准确性。在一项实验中，研究人员要求参与者按顺序向前、向后或不按特定顺序回忆去看医疗服务提供者的经历，并将结果与提供者的记录进行比对。结果表明，“自由”回忆略优于按顺序向前或向后回忆，尽管总体而言，受访者少报了20%的就诊次数；参与者的性别和自我报告的健康状况等因素也会影响回忆的质量。该实验支持了这样的观点，即应用于调查研究的认知科学方法可以提高人口调查数据的准确性。
